Gene Expression Classifier Test: Thyroid Nodule Size Irrelevant

Miriam E Tucker

July 20, 2016

The size of a thyroid nodule does not appear to affect gene expression classifier (GEC) test performance, a new study suggests.

GEC, or molecular, testing is a relatively new technology introduced to try to characterize the risk of malignancy in thyroid nodules that are deemed "indeterminate" following ultrasound and biopsy.

And there has been a concern, from prior studies, suggesting potential sampling error with use of GEC tests in nodules larger than 4 cm, explain James X Wu, MD, and colleagues at the University of California, Los Angeles (UCLA) David Geffen School of Medicine, in their article published in the July 2016 issue of Thyroid.

But in the 231 patients with 245 indeterminate nodules that they studied, all from a single tertiary referral center, the negative predictive value of GEC testing did not differ by nodule size.

"I think this is somewhat reassuring that if you have a patient with a larger nodule with a negative molecular test and the patient prefers to avoid surgery and has no other risk factors, it's something you can discuss with them," senior investigator Masha J Livhits, MD, told Medscape Medical News.

She stressed, however, that although the study shows that a benign GEC test provides some reassurance regardless of nodule size, other factors still need to be considered in deciding whether to send the patient for surgery.

"There's still a lot of clinical judgment based on what the ultrasound looks like. If it looks suspicious, it doesn't matter what the molecular testing result is, or if it's a rapidly enlarging nodule or a patient with personal risk factors....The clinical judgment is very important, maybe more important even than the molecular testing results."

In an email, Richard Kloos, MD, senior medical director (endocrinology) at Veracyte, which makes the GEC test, Afirma, that was employed in the study, told Medscape Medical News: "These findings are important and build on the robust evidence supporting use of the Afirma GEC, which has been demonstrated in a prospective, blinded, multicenter clinical validation study published in The New England Journal of Medicine, as well as in nearly 20 published clinical utility studies to date. The latter include three long-term outcome studies that followed Afirma-benign patients for at least 13 months and up to 3 years."

For Indeterminate Nodules Undergoing GEC, Size Does Not Matter

The authors explain that thyroid nodules are very common, and ultrasound with subsequent fine-needle aspiration (FNA) biopsy are currently accepted as initial diagnostic tests to identify the minority of nodules that are malignant.

But up to one-third of results from these tests are indeterminate, meaning this subset of patients often undergoes surgery to avoid a missed cancer diagnosis. Recently, GEC testing has been validated to further characterize the risk of malignancy in these so-called indeterminate nodules — those with benign GEC results can be observed whereas those with positive results generally proceed to surgery.

In their study, Dr Wu and colleagues examined 245 indeterminate nodules (from a total 2382 thyroid FNAs performed at UCLA during 2012–2015); results of GEC testing were suspicious in 132 (53.8%) and benign in 113 (46.2%).

The malignancy rate among all indeterminate nodules was 25.6%, with a GEC test sensitivity of 95.2% and specificity of 60.1%. The malignancy rate among GEC benign or GEC suspicious lesions did not vary significantly by patient age, sex, nodule size, or prebiopsy thyroid-stimulating hormone.

Positive predictive value of the GEC was consistent across nodule size (ranging from 45.5% for nodules < 1 cm to 50.0% for nodules > 4 cm; P = .29). The negative predictive value ranged from 93.3% to 100% and was also unaffected by nodule size (P = .69).

And in multivariable analysis controlling for age, sex, thyroid-stimulating hormone, and Bethesda diagnostic category, nodule size was not a significant predictor of malignancy in GEC suspicious nodules.

Prior to the advent of molecular profiling, several studies had shown a high false negative rate for cytology of large thyroid nodules and that those greater than 4 cm in particular had greater malignancy rates than smaller nodules.

Both findings had been attributed to sampling error; that is, failing to sample the area with the cancer within heterogeneous nodules.

However, not all studies demonstrated this phenomenon, the authors point out.

They hypothesize that the stability of GEC results found across nodule sizes in the current study might be because all thyroid FNAs were performed under ultrasound guidance and also that nucleic acid amplification used in the GEC test may allow for detection of smaller quantities of abnormal transcripts.

Hurthle-Cell Hurdle

This study also yielded another clinically important finding that thus far has only been reported once previously: the fact that GEC testing was far less specific in the 31 Hurthle-cell predominant nodules (HCNs) identified (13% of the total).

The presence of Hurthle-cell predominance was the only clinical factor that influenced GEC test performance and was associated with an increased rate of GEC-suspicious results despite a relatively low malignancy rate: 77.4% of HCNs had GEC-suspicious results vs just 54.0% of non-HCNs (P < .01), but the actual malignancy rate was 22.6%, similar to the 25.6% in non-HCNs.

These findings confirm those of one prior study of GEC test performance in 72 patients with HCNs, finding that a suspicious GEC did not increase the probability of malignancy in HCNs, with a specificity of only 7.5% (Thyroid. 2015;25:789-796).

In that study and the current one, a GEC benign result was almost always a true negative and could allow the patient to avoid surgery, but that was of minimal help because most GEC results for HCNs were suspicious.

So although the majority of HCNs test suspicious on GEC, most turn out to be benign.

This suggests that GEC testing may not be cost-effective in nodules identified as HCN on biopsy, Dr Wu and colleagues say.

"This is important for clinicians to know, so when we get the results we can more appropriately counsel the patient....I can't imagine that [routine] GEC [testing] can be cost-effective in Hurthle cells. The test is quite expensive and probably doesn't add that much," Dr Livhits commented.

But Dr Kloos noted, "When considering any cost-effectiveness assessment for the Afirma GEC, it is important to note that Singer and colleagues recently used a national database of health plan claims and found that the total cost of thyroid surgery and related clinical follow-up for 6 months postsurgery was over $21,000 per surgery [Curr Med Res Opin. 2016;32:1225-1232].

"This is significantly higher than any data that have been used previously to model cost savings from the test," he added.

Don't "Look at the Test Results in a Vacuum"

Dr Livhits cautioned, however, that this study shares a limitation with most real-world studies that have been published since Veracyte's original data: follow-up times too short to see whether patients with negative GEC results who don't have surgery eventually develop cancer.

"We try to follow patients for at least 6 to 12 months, but that's probably not sufficient — cancer wouldn't grow in that amount of time."

In all, she advised clinicians to know the specific performance characteristics of these molecular tests in each of their individual patients, including the underlying malignancy rates in their institutions, while keeping in mind the patient's risk factors and "not looking at the test results in a vacuum and making a decision based just on the test, but looking at the individual patient...and using the test adjunctively together with the entire clinical picture."

Dr Livhits has reported no relevant financial relationships. Dr Kloos is an employee of Veracyte.

Thyroid. 2016;26:916-922. Abstract

For more diabetes and endocrinology news, follow us on Twitter and on Facebook.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.